CN112347573A - A multi-factor efficient optimization design method for the electric field on the surface of the motor bar - Google Patents
A multi-factor efficient optimization design method for the electric field on the surface of the motor barDownload PDFInfo
- Publication number
- CN112347573A CN112347573ACN202011059259.5ACN202011059259ACN112347573ACN 112347573 ACN112347573 ACN 112347573ACN 202011059259 ACN202011059259 ACN 202011059259ACN 112347573 ACN112347573 ACN 112347573A
- Authority
- CN
- China
- Prior art keywords
- electric field
- bar
- data
- length
- maximum
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/10—Geometric CAD
- G06F30/17—Mechanical parametric or variational design
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/23—Design optimisation, verification or simulation using finite element methods [FEM] or finite difference methods [FDM]
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/12—Computing arrangements based on biological models using genetic models
- G06N3/123—DNA computing
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/12—Computing arrangements based on biological models using genetic models
- G06N3/126—Evolutionary algorithms, e.g. genetic algorithms or genetic programming
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biophysics (AREA)
- Evolutionary Computation (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Geometry (AREA)
- Software Systems (AREA)
- Data Mining & Analysis (AREA)
- Biomedical Technology (AREA)
- Molecular Biology (AREA)
- Mathematical Physics (AREA)
- General Health & Medical Sciences (AREA)
- Computational Linguistics (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Computing Systems (AREA)
- Computer Hardware Design (AREA)
- Genetics & Genomics (AREA)
- Computational Mathematics (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- Medical Informatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physiology (AREA)
- Tests Of Circuit Breakers, Generators, And Electric Motors (AREA)
Abstract
Translated fromChineseDescription
Technical Field
The invention relates to the field of simulation analysis of the performance of a large generator bar, in particular to a multi-factor optimization design method for a surface electric field of a motor bar.
Background
With the continuous development of the industry in China, the demand of the society on electric power is greatly increased, and the demand on large-scale generators is directly increased. The operation stability of the generator is greatly influenced by insulation, and particularly, an obvious electric field concentration phenomenon appears at the end part of a stator bar of the generator, and is mainly represented by corona discharge. In order to make the electric field distribution on the surface of the motor bar uniform, various influencing factors, such as the length of each section of anti-corona layer, the resistivity of each section of anti-corona layer material and the like, need to be considered during design, and the correct adjustment of the matching among the various factors is the key for improving the insulation performance of the motor bar.
In the production process of the stator bar of the generator, the motors with different installed capacities need to be designed in a targeted mode. The insulation performance of the winding bar is influenced by various factors, such as selection of main insulation and anti-corona layer materials of the motor winding bar, the length of each segment of the segmented anti-corona structure, the lap joint length between the two segments, the thickness of the main insulation and the anti-corona layer, and the like.
Aiming at the problems, the invention provides a multi-factor optimization design method for the surface electric field of the motor bar, and the multi-factor optimization is carried out on the motor bar through a composite algorithm combining a neural network and a genetic algorithm, so that the modeling time is short, the material cost of equipment is saved, and the risk and the error of manual operation are reduced.
Disclosure of Invention
The invention aims to provide a multi-factor optimal design method for a surface electric field of a motor winding bar, which is used for obtaining a multi-factor optimal matching scheme of the motor winding bar design and solving the problems of high cost, high risk, long period, low efficiency and the like of an entity test.
The multi-factor optimization design method for the surface electric field of the motor bar comprises the following steps:
(1) the method comprises the steps of collecting existing stator bar structure distribution data and the maximum tangential electric field intensity of the surface of a corresponding bar as a training data set, and reserving 10 groups of data as a test set of model accuracy. The distribution data of the stator bar is the length of each section of the anti-corona layer, the conductivity of the anti-corona layer, and the corresponding performance evaluation data is the maximum tangential electric field intensity on the surface of the bar. The relationship between the maximum tangential electric field intensity and the length and the conductivity of the anti-corona layer can be obtained by a finite element simulation result of the end part of the motor wire rod.
(2) A random number function randderm in matlab is used to generate a random number sequence for randomly sampling the training set samples.
(3) Creating a network structure according to the BP neural network, and calling to create a network net (newff (p _ train, t _ train, n), wherein the p _ train distributes data for the stator bar structure in the selected data set, the t _ train is the maximum tangential electric field intensity in the selected data set, and the n is the number of hidden layers. And a three-layer network structure is arranged, wherein 12 layers are arranged on the hidden layer, so that the model precision can be effectively improved.
(4) And (3) calling a matlab neural network tool box train (net, p _ train, t _ train) network to train by using the training set data randomly selected in the step (2), wherein the net is the initial neural network created in the step (3). The trained neural network model can enable the distribution data of the anti-corona layer of the motor bar and the corresponding maximum tangential electric field intensity to establish a multi-input single-output mapping relation.
(5) Preventing corona according to each section collected in (1)And taking the maximum value and the minimum value of the anti-corona layer data of each section as a data range. Since the conductivity changes are between orders of magnitude, log is taken of all conductivity data10. Converting the logarithm of the maximum value of the length of each anti-corona layer and the maximum value of the conductivity into binary numbers, summing the binary numbers to serve as the DNA sequence length of the individuals in the genetic algorithm, and generating 100 individuals with the DNA length by using a rand function to form an initial population of the genetic algorithm.
(6) And calling the trained neural network structure by using matlab statements sim (net, E) as a fitness function of the genetic algorithm, wherein the output is the maximum tangential electric field intensity corresponding to E, and the fitness function is used for calculating the fitness value of all individuals in the population, wherein X is ═ decimal anti-corona layer length decimal anti-corona layer resistivity.
(7) Selecting the current population according to a fitness function, eliminating individuals with low fitness (namely individuals with low evaluation on the performance of the wire rod), reserving excellent individuals, exchanging binary numbers of the same position in DNA of two adjacent excellent individuals as a cross process, and randomly negating or reserving the binary number of each position as a variation process, thereby generating the next individual.
(8) And (5) repeating the operation of the step (7) until individuals with the maximum electric field intensity below 1.5kV/cm are generated, and ending the genetic algorithm.
The invention has the beneficial effects that: 1. the advantages of the neural network are fully utilized, and the function relationship which is difficult to quantify between various design factors and performance requirements is established by using a network form; 2. the invention overcomes the defects of high cost, long period, low efficiency and the like of the entity test;
drawings
FIG. 1 is a flow chart of a multi-factor optimization design method for the surface electric field of a motor bar;
FIG. 2 is a graph comparing a neural network test set with predicted results;
FIG. 3 is a comparison of the electric field tangential electric field distribution of the preferred and other embodiments;
FIG. 4 is a three-dimensional electric field distribution diagram of the preferred optimization scheme;
FIG. 5 is a three-dimensional potential distribution diagram of the preferred optimization scheme;
FIG. 6 is a design of a randomly selected 6 sets of wire rod structures;
figure 7 is an algorithm derived optimum optimization scheme for a motor bar.
Detailed Description
The invention is further illustrated by the following examples, which are not to be construed as limiting the invention.
Fig. 1 shows a flowchart of a multi-factor optimization design method for a surface electric field of a motor bar, which is described by taking an example of optimization of an end part structure of a stator bar of a hydraulic generator with a rated voltage of 24kV and a rated capacity of 1000MW as an example:
step S1, selecting an optimization objective, obtaining an initial training data set:
the optimization target in the embodiment is to adjust the medium resistance length, the medium resistance resistivity and the high resistance resistivity of the corona prevention layer of the motor bar so that the maximum tangential electric field on the surface of the motor bar is lower than 1.5 kV/cm. Establishing a sample set (X) by combining a finite element simulation model of the end part of the electrode wire rodi，Yi) 1, 2., N }, where X represents an input item of the mapping, Y represents an output item of the mapping, and the input items are medium resistance length, medium resistance resistivity, and the output item is electric field strength. The sample size N is 100, with 90 sets being the training data set and 10 sets being the test set.
Step S2, normalization processing:
because the magnitude of the selected training set parameters is greatly different, the training set and the test set are normalized, and the normalization processing formula is as follows:
step S3, establishing a neural network structure:
the MATLAB neural network tool box is set into a three-layer network structure comprising an input layer, a hidden layer, a data layer and a data layer,And outputting the layer, wherein the hidden layer is provided with 12 nodes. The training times can be adjusted according to the accuracy of the training result, the initial setting is 1000 times, the learning rate is set to 0.01, and the training mean square error is set to 10-5。
Step S4, training the network, evaluating the accuracy of the neural network:
training a neural network by using a MATLAB tool box, calling to create a netword net (X, Y,12), inputting normalized training set data into the neural network, and calling a MATLAB neural network tool box train (net, X, Y) network for training. And (3) putting 10 groups of input items in the test set into the neural network, comparing the output obtained by calculating the neural network with the electric field intensity corresponding to the test set, and verifying the accuracy of network calculation, wherein the calculation accuracy of the neural network is higher as can be seen from fig. 2.
Step S5, initializing the genetic algorithm initial population:
and taking the maximum value and the minimum value of each section of anti-corona layer data as a data range according to the data of the medium resistance length, the medium resistance resistivity and the high resistance resistivity of 100 groups of motor bar anti-corona layers in S1. Since the conductivity changes are between orders of magnitude, log is taken of all conductivity data10. Converting the logarithm of the maximum value of the length of each anti-corona layer and the maximum value of the conductivity into binary numbers, summing the binary numbers to serve as the DNA sequence length of the individuals in the genetic algorithm, and generating 100 individuals with the DNA length by using a rand function to form an initial population of the genetic algorithm. The population size is set to 100.
Step S6, constructing a fitness function, and evaluating individual fitness:
using matlab statement sim (net, E) to call the trained neural network structure as the fitness function of the genetic algorithm, wherein the output is the maximum tangential electric field intensity corresponding to X, and the maximum tangential electric field intensity is used for calculating the electric field intensity corresponding to all individuals in the population, wherein E is [ length of middle-resistance anti-corona layer, length log of middle-resistance anti-corona layer, and length log of high-resistance anti-corona layer10(medium resistivity) log10(middle and high resistivity)]. Outputting the electric field intensity according to the parameter data carried by the individual, and the individual fitness with smaller electric field intensityHigher. In order to be matched with the use of the neural network, when the fitness of an individual is calculated, each segment of the DNA binary code of the individual is converted into decimal data, and then an sim statement is input.
Step S7, genetic manipulation, generating next generation individuals:
based on individual fitness, selecting excellent individuals, eliminating individuals with low fitness, exchanging binary numbers of two adjacent excellent individuals at the same position to be used as a cross process, and randomly negating or reserving the binary number at each position to be used as a variation process, thereby generating the next individual.
Step S8, generating an optimization result:
and repeating the steps S7 and S8 until the population of a certain generation obtains the maximum fitness, namely the minimum electric field strength meets the optimization standard, ending the genetic optimization process, and obtaining the data of the middle resistance length, the middle resistance resistivity and the high resistance resistivity of the corona prevention layer of the optimal motor bar and the corresponding maximum electric field strength.
The invention can obtain the optimal optimization scheme of the motor bar structure design, randomly extracts 5 groups of motor bar structure design schemes (see figure 6), and compares the optimal scheme with the optimization scheme (see figure 7), and the maximum electric field of the optimization result is the lowest value as can be seen from figure 3, thus proving that the optimization effect is better. Fig. 4 and 5 show simulation diagrams of the tangential electric field and the electric potential of the surface of the bar corresponding to the optimal optimization scheme, and it can be seen that the uniformity of the electric field distribution is better and the maximum electric field intensity is lower.
In the description herein, references to the terms "one embodiment" or "an embodiment" mean that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. In this specification, the schematic representations of the terms used above do not necessarily refer to the same embodiment or example.
The above examples represent several embodiments of the present invention, and the description thereof is more specific and detailed, but not to be construed as limiting the scope of the present invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the inventive concept, which falls within the scope of the present invention. Therefore, the protection scope of the present patent shall be subject to the appended claims.
Claims (2)
1. The multi-factor optimization design method for the surface electric field of the motor bar is characterized by comprising the following steps of:
(1) acquiring distribution data of an existing stator bar structure and the maximum tangential electric field intensity of the surface of a corresponding bar as a training data set, and reserving 10 groups of data as a test set of model precision, wherein the distribution data of the stator bar is the length of each section of an anti-corona layer, the conductivity of the anti-corona layer, and the corresponding performance evaluation data is the maximum tangential electric field intensity of the surface of the bar, the relationship between the maximum tangential electric field intensity and the length and the conductivity of the anti-corona layer can be acquired by a finite element simulation result of the end part of a motor bar (2) a random number function randderm in matlab is used for generating a random digital sequence for randomly sampling a training set sample (3) to create a network structure according to a BP neural network, and creating network words net = newff (p _ train, t _ train, n) are called, wherein p _ train is the distribution data of the stator bar structure in the selected data set, and t _ train is the maximum tangential electric field intensity in the selected data set, n is the number of hidden layers, a three-layer network structure is arranged, 12 layers are arranged on the hidden layers, the model precision can be effectively improved (4), training is carried out by calling a matlab neural network tool kit train (net, p _ train, t _ train) network by utilizing training set data randomly selected in the step (2), wherein the net is an initial neural network established in the step (3), the trained neural network model can enable distribution data of a motor bar anti-corona layer and corresponding maximum tangential electric field intensity to establish a multi-input single-output mapping relation (5), according to the length of each anti-corona layer and each section of electric conductivity collected in the step (1), the maximum value and the minimum value of each section of anti-corona layer data are taken as a data range, and because the change of the electric conductivity is changed among a plurality of orders of magnitude, log is taken for all the electric conductivity data10Converting the logarithm of the maximum length value and the maximum conductivity value of each anti-corona layer into binary number, and summing the binary number as the number in the genetic algorithmDNA sequence length of the body, generating 100 individuals with the DNA length by using a rand function, forming a genetic algorithm initial population (6), calling a trained neural network structure by using matlab statements sim (net, E) as a fitness function of the genetic algorithm, outputting the maximum tangential electric field intensity corresponding to E, and calculating the fitness value of all individuals in the population, wherein X = [ decimal anti-corona layer length decimal anti-corona layer resistivity = [ decimal anti-corona layer length ] is](7) And (3) selecting the current population according to a fitness function, eliminating individuals with low fitness (namely individuals with low evaluation on the performance of the wire rod), reserving excellent individuals, exchanging binary numbers at the same position in DNA of two adjacent excellent individuals to be used as a cross process, and then randomly negating or reserving the binary number at each position to be used as a variation process, so as to generate the next individual (8), repeating the operation of the step (7) until the individuals with the maximum electric field intensity below 1.5kV/cm are generated, and ending the genetic algorithm.
2. The multifactor optimal design method for the surface electric field of a motor bar as in claim 1, wherein a reference design can be given to achieve theoretical maximum performance based on different performance and bar design requirements while ensuring accuracy.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011059259.5ACN112347573A (en) | 2020-09-30 | 2020-09-30 | A multi-factor efficient optimization design method for the electric field on the surface of the motor bar |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011059259.5ACN112347573A (en) | 2020-09-30 | 2020-09-30 | A multi-factor efficient optimization design method for the electric field on the surface of the motor bar |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112347573Atrue CN112347573A (en) | 2021-02-09 |
Family
ID=74361409
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011059259.5APendingCN112347573A (en) | 2020-09-30 | 2020-09-30 | A multi-factor efficient optimization design method for the electric field on the surface of the motor bar |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112347573A (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104269953A (en)* | 2014-09-24 | 2015-01-07 | 湖南科技大学 | High-voltage motor stator coil corona-preventing structure optimizing method based on electric field analysis |
| CN104899375A (en)* | 2015-06-05 | 2015-09-09 | 株洲时代新材料科技股份有限公司 | Main insulating anti-corona structure optimization calculating method used for high-voltage motor |
| CN106919773A (en)* | 2017-04-25 | 2017-07-04 | 哈尔滨理工大学 | A kind of numerical computations APP for big stator winding conductor bars in electrical machines end model |
| CN109165764A (en)* | 2018-06-26 | 2019-01-08 | 昆明理工大学 | A kind of line loss calculation method of genetic algorithm optimization BP neural network |
| US20200133256A1 (en)* | 2018-05-07 | 2020-04-30 | Strong Force Iot Portfolio 2016, Llc | Methods and systems for sampling and storing machine signals for analytics and maintenance using the industrial internet of things |
- 2020
- 2020-09-30CNCN202011059259.5Apatent/CN112347573A/enactivePending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104269953A (en)* | 2014-09-24 | 2015-01-07 | 湖南科技大学 | High-voltage motor stator coil corona-preventing structure optimizing method based on electric field analysis |
| CN104899375A (en)* | 2015-06-05 | 2015-09-09 | 株洲时代新材料科技股份有限公司 | Main insulating anti-corona structure optimization calculating method used for high-voltage motor |
| CN106919773A (en)* | 2017-04-25 | 2017-07-04 | 哈尔滨理工大学 | A kind of numerical computations APP for big stator winding conductor bars in electrical machines end model |
| US20200133256A1 (en)* | 2018-05-07 | 2020-04-30 | Strong Force Iot Portfolio 2016, Llc | Methods and systems for sampling and storing machine signals for analytics and maintenance using the industrial internet of things |
| CN109165764A (en)* | 2018-06-26 | 2019-01-08 | 昆明理工大学 | A kind of line loss calculation method of genetic algorithm optimization BP neural network |
Non-Patent Citations (5)
| Title |
|---|
| 候海波: "大电机主绝缘端部防晕结构优化与试验研究", 中国优秀硕士学位论文全文数据库工程科技II辑, pages 30 - 43* |
| 杨林军;: "发电机局部放电非线性模式识别现状及展望", 现代电子技术, no. 14, pages 116 - 120* |
| 田质广等: "基于遗传算法的神经网络在发电机定子超高频局部放电模式识别中的应用", 大电机技术, no. 01, pages 36 - 40* |
| 蒋雄伟等: "基于遗传算法的神经网络在局部放电模式识别中的应用", 西安交通大学学报, vol. 33, no. 12, pages 1 - 4* |
| 高凯等: "基于散点集分形特征的局部放电模式识别研究", 中国电机工程学报, vol. 22, no. 05, pages 22 - 26* |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108876054B (en) | Short-term power load forecasting method based on improved genetic algorithm to optimize extreme learning machine | |
| CN102768701B (en) | High-voltage switch cabinet insulator electric field optimization method based on quantum genetic algorithm | |
| CN113158573B (en) | An optimal power flow acquisition method for regional distribution network of small hydropower clusters based on deep learning | |
| CN107612016B (en) | Planning method of distributed power supply in power distribution network based on maximum voltage correlation entropy | |
| CN105574266B (en) | A kind of disc insulator based on Multiple-population Genetic Algorithm electrically and mechanically performance synthesis optimum design method | |
| CN112684346B (en) | Lithium battery state of health estimation method based on genetic convolutional neural network | |
| CN107506914B (en) | A dynamic expansion planning method for substations considering changes in the penetration rate of distributed power sources | |
| CN113095648A (en) | Power distribution network fault power failure risk index evaluation method and system based on data driving | |
| CN111105005B (en) | Wind power prediction method | |
| CN114065551A (en) | Probabilistic power flow calculation method for AC-DC hybrid grid based on Copula | |
| CN105186499A (en) | Multi-target probabilistically optimal power flow fuzzy modelling and solving method for power distribution network | |
| CN111859807A (en) | Initial pressure optimizing method, device, equipment and storage medium for steam turbine | |
| CN110059891A (en) | A kind of photovoltaic plant output power predicting method based on VMD-SVM-WSA-GM built-up pattern | |
| CN107994582A (en) | Method and system for reconstructing power distribution network containing distributed power supply | |
| CN112347573A (en) | A multi-factor efficient optimization design method for the electric field on the surface of the motor bar | |
| CN106953343B (en) | A kind of phase three-wire three uncompensated load only utilizes the compensation method of capacitive element | |
| CN109376494B (en) | Probability power flow determination method and system based on power system | |
| CN115221787B (en) | Multi-objective optimization method and system for cylindrical permanent magnet linear motor based on NSGA-Ⅱ | |
| CN112765842B (en) | Optimization design method for voltage equalizing structure of combined insulator | |
| CN116522835A (en) | Reconfigurable battery system topology structure optimization design method | |
| CN115986811A (en) | Photovoltaic station model parameter identification method based on genetic algorithm | |
| CN113346501A (en) | Power distribution network voltage optimization method and system based on brainstorming algorithm | |
| Yasin et al. | Multiobjective quantum-inspired evolutionary programming for optimal location and sizing of distributed generation | |
| CN110782080A (en) | A Power System Structure Optimization Method Based on Population Performance Ranking Selection | |
| CN104699950A (en) | Method for calculating stochastic load flow with correlation problem of stochastic power units |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication | Application publication date:20210209 |